Tag Archives: midterm elections

Second Term Worse than the First

Posted on by

Consumer sentiment is far worse in 2nd term than in 1st

Analysts of President Donald Trump’s second term, and the outlook for the midterm elections on Nov. 3, have reasonably focused on Trump’s job approval. After holding above the first term approval trend, the second term approval fell below the first term in January and has recently fallen more after the start of the Iran war.

This is well known and I have nothing to add.

What is less often considered are opinions of the economy and especially the comparison of first and second term opinion. This deserves more attention.

The University of Michigan’s index of consumer sentiment is a long-running monthly survey of how Americans feel about the economy. As an index, high values reflect optimism and positive feelings, low values show pessimism and negative feelings. An index value of 100 (where the index stood in 1966) reflects quite positive views of the economy including current conditions and future expectations. As an index, values can be above or below 100, i.e. this is not a percentage.

On April 10, the Michigan survey reported consumer sentiment at 47.6, a record low in polling dating back to 1952. Four of the 5 lowest values ever have come in the last 4 years, with two in the Biden administration in June and July 2022 at 50.0 and 51.5 respectively (at the peak of the inflation surge), and in November 2025 with an index at 51.0, in addition to the current all time low. The 5th lowest index ever was 51.7 in May 1980. In short, despite objective measures of GDP, unemployment and inflation having been far worse in some earlier years, Americans are stunningly sour on the economy,

The comparison of consumer sentiment in the first Trump administration and in the second is the point of this post. The chart highlights the first term up to the 2018 midterms and the second term so far. The average sentiment in the first 23 months of term 1 was 97.5. The average so far in term 2 is 55.5, with the most recent reading at 47.6. That is a 42 point drop from average to average and a 49.9 point drop from average to current reading.

To state the obvious: economic sentiment was a tremendous advantage in the first Trump term and is a tremendous burden in the second.

Sentiment plummeted when the Covid pandemic arrived in early 2020, then began to recover into 2021 before the spike in inflation in the second half of 2021 drove sentiment to the then all-time low of 50 in June 2022. Sentiment recovered somewhat through most of the 2nd half of the Biden administration though it dipped in the run-up to the 2024 election. That persistent negative view of the economy was a constant weight on Biden’s support and ultimately on Harris’ vote.

During Trump’s second term the trend has been sharply down from a peak of 74.0 in December 2024 immediately after his reelection, to 64.7 in the first month of the new term with irregular month to month movements and an overall downward trend.

The low consumer sentiment index means the economy is virtually guaranteed to remain the top concern for voters, and therefore the issue all candidates have to discuss (and claim to fix, with more or less persuasiveness). Above all, this economic gloom will be the atmosphere of the election.

Whatever Trump’s approval rating was in the first term, he could count on an electorate optimistic about the economy. In the second term economic pessimism can only be a drag on his approval and the fortunes of the Republican party in November.

Low consumer sentiment doesn’t guarantee big GOP seat losses in November. While it is correlated with seat loss the fit is quite loose. Presidential approval is a better predictor of seat loss. But economic concerns write the script for the 2026 election.

Let’s do take a moment to reiterate what all have said before: compare Trump approval in term 1 and term 2. While there has been steady decline in net approval in the second term, the first year of the first term was lower and reached what is still the lowest point of either term at -19.4. My estimate of the lowest net approval of the second term is -16.8 on April 6. As of April 9 the net approval estimate is -16.5. You can, of course, consult the many other websites on Substack and elsewhere for alternate estimates of the approval trends. These are mine. Some are a little better for Trump and some a little worse. We all tell the same qualitative story and show very similar bumps and wiggles.

I’ve added annotation for some significant events around the times of movement in approval. See Elliott Morris’s look at consequential events for Trump 2 approval at his Strength in NumbersSubstack. I have slightly different notable events based on my judgement rather than statistical fits.

In the first term approval fell around Trump’s effort to replace Obamacare, and continued down until the passage of the Ryan tax cut package in December 2017. After that, net approval rose until it stabilized at about -10 where it remained through the midterms.

In the second term so far, we’ve not seen a sustained recovery in net approval. After the negative reaction to “Liberation Day” tariffs, approval declined until Trump announced he was pausing the tariffs and negotiating. That bought back some approval points in the late spring, only to again start declining by June. Likewise the October government shutdown coincided with a drop in approval, with some recovery after the shutdown ended. The most recent period of decline has not seen a similar recovery period so far.

This is not to say there can be no approval recovery. We know not what events may occur over the next 207 days until the election. But we do know that inflation and the cost of living has been the most important problem in surveys since the inflation spike in 2021-22, and it has remained the number one problem throughout Trump’s second term. The vast difference in consumer sentiment in the second term compared to the first shows vividly that the economy is not the life preserver the president and his party seek.

The paradox that an unpopular Democratic party keeps winning

Posted on by

Democrats have reservations about their party but unite in opposing Trump

The Democratic party is less popular than the Republican party in both Wisconsin and national Marquette Law School polls since Jan. 2025. Yet Democrats keep strongly over-performing in both general and special elections in 2025 and 2026. Most recently Democrats flipped the Florida legislative seat that includes Mar-a-Lago, a district that went +11 for Trump in 2024 but went +2 for the Democratic legislative candidate on March 24. How can this paradox be explained?

I previously looked at this in my national polling here. Today let’s look at the most competitive state in the nation since 2016, Wisconsin. Data are from Marquette Law School polls of registered voters in October 2025, and February and March 2026.

Both parties are underwater, but Democrats more so

Both parties have net negative favorability ratings, meaning more have unfavorable than favorable views of each party. Since October there has been little change in overall favorability for each, with Democratic party favorability more net negative than for the Republican party. GOP net favorability has been around -10 points while Democratic net favorability has been twice as large, in the -20s.

Where does this added negativity to the Democrats come from? Not from the opposite party. Republicans have extremely negative views of Democrats and Democrats return the favor with equally net negative views of Republicans. Likewise this difference between parties doesn’t come from independents, who strongly dislike both parties though they give Republicans an 8 point more net negative rating than they give Democrats.

The partisan gap comes in feelings about voters’ own party. Republicans give the GOP a net positive +74 point rating, but Democrats give their party just a +56 point rating. Our partisans hate the other party equally but don’t love themselves equally.

As in horror movies, “the call is coming from inside the house.”

The same result is seen in my national polling where we also ask about approval of the job Democrats in Congress and Republicans in Congress are doing. Partisans strongly disapprove of the other congressional party but Democrats are less approving of congressional Democrats while Republicans are more approving of their party in Congress. Democratic discontent applies equally to favorability of the party in general and specifically to the job performance of the congressional party.

This asymmetry in partisan views take a different turn when the question is about President Donald Trump. Across the three polls since October Trump’s overall approval is 44% and disapproval is 54%, for a net rating of -10 percentage points. Here, though, Democrats are stunningly united in disapproval. Republicans strongly approve but not so much as Democrats disapprove. And independents are also strong in their disapproval of the president. This negative balance, seen in both Wisconsin and national Marquette polls, explains why Republican candidates have suffered in elections in 2025 and 2026 despite Trump winning the presidency in 2024.

Democrats are less thrilled with their party but virtually all are agreed in their dislike for Trump. This, plus strong independent dissatisfaction with Trump, has powered Democratic over-performance and wins in the 2025 general elections in Virginia and New Jersey and in special elections throughout the country.

Where dissatisfaction comes from within each party

Which Democrats are less satisfied with their party, and where does Republican dissatisfaction arise in the GOP?

Very few Republicans consider themselves to be liberal or very liberal, and very few Democrats describe themselves as conservative or very conservative. I lump these few outliers in with moderates in each party. (To see how that has changed over the past 15 years in Wisconsin see my earlier post here.)

Here we look at very conservative to moderate Republicans and at moderate to very liberal Democrats and their favorability to their own parties.

In the GOP those who are very conservative have the highest net positive feelings for the Republican party, and those who are conservative are almost as positive. Where there is less positivity is with moderate Republicans, who remain solidly positive but considerably less so than either type of conservative.

For the Democrats, those who are very liberal are the least positive to the party, just +24 points net favorable, while liberals are most positive and moderates pretty positive but not enormously so. Comparing Democrats with Republicans, very liberal Democrats are much less favorable to the Democratic party than very conservative Republicans are to their party. And liberal Democrats are less positive than conservative Republicans to their respective parties. Only among moderates does the Democratic party enjoy a modest advantage in net favorability, a +56 vs +45 among Republicans.

The asymmetry between parties is not only with one ideological camp, but the strikingly low favorability among very liberal Democrats reflects the ongoing debate within the party between those urging moderation and those pushing for a more clearly liberal or progressive party.

Some of these discrepancies are likely due to the natural frustration of a minority party unable to control the congressional agenda and able at best to block or stall legislation in the Senate. But the differences by ideology also point to a divide over the direction of the Democratic party.

These are the things that explain why the Democratic party has consistently less favorable ratings than the Republican party, and that the discrepancy is substantially due to differences within the parties.

The paradox unriddled

The paradox of strong electoral performance by the less popular party is driven by opposite forces. Democrats differ about themselves but they nearly unanimously oppose Donald Trump, and they are joined by solid opposition to Trump among independents. Republicans have maintained very strong support for Trump within their party, but not so strong as Democratic opposition and with quite weak appeal among independents.

Democrats may not agree on what they are for, and are frustrated by their party’s inability to block Trump, but they absolutely know what (or who) they are against. This has provided a strong electoral advantage when Republican candidates across the country have so closely tied themselves to President Trump, and where dissent brings presidential condemnation and primary challenges. November is seven months away and the details of candidates and issues are not yet certain, but it is against this background that the less popular party nonetheless enjoys a midterm advantage.

House Retirements and Targeted Districts

Posted on by

While a substantial number of members of the House of Representatives are retiring, don’t expect these retirements to produce many flipped seats or shifts in the ideological makeup of either party.

As of January 13, 47 members of the House have announced their retirement, 21 Democrats and 26 Republicans. (I’m not counting resignations by Majorie Taylor Greene and Mikie Sherrill whose seats will be filled with special elections this year.)

The retirement rate has been running a bit ahead of recent cycles as of this date, which were 42, 34, 41, and 40 from 2018 to 2024. Still, I don’t think we are seeing extraordinarily high levels of retirements, as some commentary suggests. In the end those previous four cycles produced totals of 52, 36, 49 and 45 retirements, suggesting we may end up in the mid-to-upper 50s this year. Past retirements are from Ballotpedia.org.

The main point I want to make here is that the retirements are spread pretty widely throughout both Republican and Democratic caucuses by ideology and 2024 vote margin. The solid dots are retiring members. These are not endangered incumbents who barely scraped by in 2024, nor are they ideological outliers relative to their caucuses.

The figure shows all House members by vote margin and by left-right ideology, using Nokken-Poole dimension 1 ideology scores from VoteView.com. These scores are based on roll call votes by the members. Nokken-Poole is a variant of the widely used Nominate scores. Nokken-Poole scores range from -0.848 for the most liberal member to 0.986 for the most conservative member. Vote margin is the percentage for the Republican candidate minus the percentage for the Democrat, so negative margins are Democratic wins and positive ones are Republican victories.

Among Republicans, the median 2024 vote margin is 28.2 percentage points, and the median for retiring Republicans is 26.1 points. On ideology, the median Nokken-Poole score is 0.542 (higher scores are more conservative.) Among retiring Republicans the median is 0.581

Democratic retirees have somewhat larger vote margins, -36.8 percentage points, than their caucus as a whole, -27.0 points. On ideology, the retiring Democrats are also more liberal, -0.461, than the full Democratic caucus, -0.394. These are modest differences, however, and the figure makes clear retirements are well scattered throughout both caucuses.

The upshot of this distribution of retirements is that it does not open up many opportunities for turnover as most retirees enjoyed reasonably secure margins in 2024. Nor are retirements likely to significantly shift the ideological balance in the House given that retirees are ideologically pretty representative of their caucuses. While open races are less predictable than incumbent ones, the strong partisan lean of most of these districts means we should expect no more than a handful of these seats to potentially flip.

DCCC and NRCC target districts

Both the Democratic Congressional Campaign Committe (DCCC) and the National Republican Congressional Committee (NRCC) have released lists of districts being targeted as pick up opportunities. Compare this figure with the retirements above. The targeted districts are, as you would expect, far more concentrated in races that were narrowly decided in 2024. (These lists were released by the NRCC on March 17, and by the DCCC on April 8. They do not include changes or additions after some states redistricted in 2025. These are the members’ districts in the 119th congress.)

Republicans on the DCCC list have a median vote margin of 6.8 percentage points, much closer than the caucus median of 28.2 points. They are also less conservative, 0.384, than the full caucus, 0.542.

Democrats on the NRCC list also had much closer 2024 races, with a median of -3.2 percentage points compared to -27.0 for the full caucus. These Democrats are also less liberal than the caucus, with a median Nokken-Poole score of -0.221 compared to the caucus median of -0.394.

If you are looking for change in the House, look at the districts each of the parties are focusing on. They have a much greater chance of flipping than the seats of retiring members, and would be more likely to remove relatively moderate members of either party. The latter fact will also contribute to polarization in the House. Rather than target ideologically extreme members of the opposition party, both Democats and Republicans target close races, which also happen to be where the most relatively moderate members are.

Midterm seat losses, 1862-2022

Posted on by

The president’s party almost always suffers in the House

Midterms are less than a year away so it’s time to look back at the record. 

In the House, the president’s party has lost seats in all but four midterms since 1862, and one of those, 1902, was a year the House expanded so the Republican gains fell short of Democratic gains that year. After 1934 it wasn’t until 1998 that the president’s party gained seats, then the rare event repeated in 2002. Not since.

This regularity over 160 years is hard to attribute to the circumstances of the moment. Likewise the hope that “this year will be different” has been a forlorn one. The size of the seat loss, on the other hand, has varied considerably and is correlated with presidential approval (Clinton in 1998 and Bush in 2002 were unusually popular, as was Roosevelt in 1934) and the state of the economy. Popular presidents lose fewer seats, unpopular ones more. Good times go with smaller losses, bad times with greater losses. 

In 2026 we will have the unprecedented circumstance of a large number of mid-decade redistricting decisions as the parties battle to see who can gerrymander the most seats to their advantage. That landscape is still being painted.

In the Senate the pattern of losses are much less dependable than in the House. The president’s party usually loses seats, and the gains have been small since the 1940s. The vagaries of which seats are up and how many for each party adds uncertainty to the Senate picture. And, of course, prior to 1913 the Senate was not elected by popular vote.

In the House, the second midterm for a president produces barely grater losses than the first midterm (an average of 28.9 in the second, versus 24.9 in the first.) There isn’t a “six-year itch” on the House side. The Senate has been a different matter, with second midterm losses averaging 6 seats rather than 2 in the first midterm (since 1946.) One plausible explanation for the Senate is that the 6th year is the reelection of a Senate class elected with the president for his first term. To the extent a winning first term president brings along some partisan companions, they run without his coattails 6 years later. The table shows the seat changes and the averages since 1946.

Next year there will be inevitable discussion of whether 2026 is Trump’s second midterm. Obviously it is in one sense. But for the Senate, this will be the class elected with Joe Biden in 2020, not the class elected with Trump in 2016. This group had mild pro-Democratic national forces either helping or hurting them in 2020. Given this difference in the election cycle, the average 6 seat loss may not be as good a representation of expectations for Senate seats in 2026.

Pre-election polling trends

Posted on by

Here is my look at the pre-election polling, as of Nov. 5, 2022.

The raw polling data is from FiveThirtyEight.com, who generously allow download of their polling database. The trends are my estimates and not those of FiveThirtyEight or anyone else.

I include both LV and RV poll results because I believe this reflects uncertainty about turnout, which most do not include in their estimates.

Note the trends as well as the point estimates. The last date of polling is shown in each chart, but most polling was completed a week or more before the election.

Charts are in order of Cook rating of the race and alphabetical within Cook rating (so Solid-D to Solid-R and alphabetical within rating group.

A race had to have at least 5 polls to be included here.

Senate races first, then governor races.

Now the Governor races

Generic Congressional Ballot Review

Posted on by

The congressional generic ballot is one of the few indicators with a long history of non-presidential vote outcomes we have. I’ll look at how good a predictor it is at another time. Here is simply a review of the trends over each election cycle. The chart above shows each cycle since 1999-2000 until the current cycle, as of Oct. 5, 2021.

And for the obsessive, here is a chart of each cycle since 1945-46. It Is unreadable without zooming in, so download it and zoom as you wish.

The .pdf version is best for zooming in. You can download it here.

Updated 2021-10-06 to correct errors in 1981-82 chart.Thanks to @SteveEvets8 for spotting the errors.

Vote Surge and Decline

Posted on by

Hmmmm. I wonder what this is about? (OK: tl;dr it is about midterm elections and what to expect in 2022).

“Surge and decline” is the title of a 1960 article by Angus Campbell:

Abstract

The tides of party voting are as fascinating as the fluctuations of economic activity. Regularities in the ebb and flow of voting with the alternation of Congressional and Presidential elections challenge the analyst to find an explanation. This article seeks it in propositions rooted in survey data.

Campbell’s explanation was that presidential elections are largely driven by “short term forces” that provide a temporary advantage to one party, usually the winner of the presidency. Then in the midterm, with no presidential election, short term forces are less important and the electorate shrinks with lower turnout (a “low stimulus election” relative to presidential) and the “long-term force” of partisanship becomes relatively more dominant in the midterm. The result is a return to the partisan balance after the surge favoring the presidential winner two years earlier.

This is a beautifully elegant theory. It is rooted in a simple model of partisanship, short-term forces and the inevitable decline of turnout in midterms. I love it.

Alas, it no longer commands general acceptance as a theory of midterm seat loss. More emphasis is now given to presidential approval, economic conditions and incumbency. Those theories bring substantial empirical evidence, and are certainly sensible. But to me they lack Campbell’s beautiful simplicity.

I’m not here to argue theories of midterm loss, but rather to simply illustrate the votes side of midterm losses. That the president’s party almost always loses house seats in midterms is a fact. Here I look at the decline in votes for the president’s party from presidential to midterm.

For fun, I’m going to walk you through the puzzle and steps that end with the figure at the top of this post. Here is the first step. What IS this??

As my poor former students know, I enjoy starting a class example with a mystery. Show the chart above, invite speculation as to what it might be. I love this one because it looks like random noise with no relationship at all.

Next step: Revealing the variables.

OK here are the variables. National Democratic percentage of the 2-party House vote in the midterm by the national 2 party vote in the previous presidential election. Not much of a relationship. Some votes go up (above the diagonal) and about as many go down (below diagonal.)

Ahh, but what about control of the presidency? The midterm-loss of seats by the president’s party is well known. The national vote, here, shows the same pattern. Dems do better (above diagonal) with GOP president, and do worse with a Dem president, almost always (except 2002.)

This is the surge and decline of votes. Almost always the president’s party wins a smaller share of votes in the midterm than they did in the presidential year. Not surprisingly, fewer votes translate into fewer seats, but that’s not our topic here. For that see this post.

The pattern is clear if we fit a regression for midterms with Rep presidents (red line) and one for Dem pres (blue line). Now the upward slope is clear (midterm performance IS related to prior pres vote) and the party of president shifts the lines up (Rep pres) or down (Dem pres).

FWIW the red and blue lines are nearly parallel. A test of a pooled model finds the difference in slopes to be statistically insignificant (p=.9465). I’m using the separate regressions here, but there would be minimal difference for a pooled model.

So what does the model tell us about, say 2018? The red line estimates the Dem 2-pty vote in 2018 to be 53.1%, up from the Dem 2-pty 2016 vote of 49.5%. In fact, Dems got 54.4%, 1.3 points better than the model and 4.9 points over their 2016 performance.

We don’t know how Dems will do in 2022, but we do know how they did in 2020 and that there is a Dem president. The vertical black line shows the actual Dem share of 2020 House vote, and the blue arrow shows the fit: a predicted 47.8% in 2022, down from 51.6% in 2020.

This is the dilemma of every presidential party: they are almost certain to lose votes in midterm elections. For closely divided congresses (looking at you 117th) this imperils majorities. 2002 was an exception, with 1998 and 1990 almost being exceptions.

How do votes translate into seats? The relationship shifted after 1994 undoing a long standing Dem advantage. The votes-to-seats model expected Dems to hold 50.6% or 220 seats in 2021 (actual post-election was 222). For 2022 the estimate is 45.0% of seats. That would be 196 seats, a loss of 26 from the post-2020 election total.

The president’s party gained house seats in 1934, 1998 and 2002, and lost share of seats in every other midterm since 1862. Reps gained seats in 1902 as the House expanded but actually lost share of seats as Dems gained more that year.

If the historical pattern applies in 2022 the Democrats are unlikely to hold control of the House. In addition there will also be the effect of redistricting. Both parties will have an incentive to gerrymander for every advantage possible where they control the process.

Of course the past pattern may change. Political skill or folly might shift the balance away from the models. The model is useful because it gives us a basis for our expectations. We can judge party performance by whether outcomes exceed or fall short of model expectations. 12/12

Midterm Seat Loss

Posted on by

I’ve been shocked to hear several sources I respect get the midterm seat loss story wrong. So here is my effort to clarify.

The president’s party almost always loses House seats, but there have been 4* exceptions since 1862: 1902, 1934, 1998 & 2002. *HOWEVER in 1902 the House expanded so while Reps gained seats Dems gained more, thus Reps won a smaller percentage of seats that year. So the presidents party has lost strength in all but 3 midterms since 1862.

In the Senate the president’s party usually loses seats, but not as reliably as in the House. There have been 6 exceptions since 1960.

There is little difference, on average, in House seat losses in 1st vs 2nd midterms. An average -26.4 in 1st and -28.1 in 2nd. NO SIX YEAR ITCH! NO 1ST MIDTERM CURSE EITHER, for that matter.

2nd midterms HAVE been worse in the Senate: -2.3 in 1st, -6.0 in 2nd.

So PLEASE stop saying the president’s party only gains seats “once in the last 100 years”– you know who you are. The right answer is “three times in the last 100 years.”

And don’t imply the Senate is as predictable as the House. They aren’t the same.

And… 1st term vs 2nd? Nah. This is another rant as many people bring up “first midterm” (and in a 2nd term almost always talk about the “second midterm”) as if that mattered. It doesn’t, on average. It does vary across presidencies with some bigger losses in 1st and some in 2nd midterm.

And will 2022 be different? I don’t know. But we should get the history right.

Data details

These seat changes reflect the immediate outcome of the November election. Sometimes members die, change party or resign before the Congress is sworn in, and of course changes can occur during the Congress.

Brookings hosts Vital Statistics on Congress. Note they have a typo for 1998 indicating a loss rather than a gain. I use them here with that fix

Here is the Vital Statistics table.

Small differences if you use the Clerk of the House table, p59